CN109495317B - Data network flow prediction method and device - Google Patents
Data network flow prediction method and device Download PDFInfo
- Publication number
- CN109495317B CN109495317B CN201811526034.9A CN201811526034A CN109495317B CN 109495317 B CN109495317 B CN 109495317B CN 201811526034 A CN201811526034 A CN 201811526034A CN 109495317 B CN109495317 B CN 109495317B
- Authority
- CN
- China
- Prior art keywords
- flow
- users
- historical
- data network
- different bandwidths
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/14—Network analysis or design
- H04L41/145—Network analysis or design involving simulating, designing, planning or modelling of a network
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/14—Network analysis or design
- H04L41/147—Network analysis or design for predicting network behaviour
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
- H04L43/0876—Network utilisation, e.g. volume of load or congestion level
Abstract
The invention provides a data network flow prediction method and a device, wherein the method comprises the following steps: acquiring total traffic historical data of a data network, historical online user numbers of different service types and different bandwidths; based on the established flow prediction model, determining historical average use flow of users with different service types and different bandwidths according to total flow historical data of the data network and historical online user numbers with different service types and different bandwidths; and predicting the total flow of the data network after capacity expansion based on preset target capacity expansion users of different service types and different bandwidths according to the historical average use flow of the users of different service types and different bandwidths. The scheme can reflect the whole traffic condition of the data network traffic and reflect the whole traffic growth trend of the data network; the user behavior of each bandwidth under the service type can be embodied, and a basis can be provided for the large speed increase of the broadband.
Description
Technical Field
The present invention relates to the field of communications technologies, and in particular, to a method and an apparatus for predicting data network traffic.
Background
The flow prediction is a key concern of each IP operator, the accurate flow prediction has great guiding significance on the expansion of the network, and the operator can be guided to develop more refined service popularization in the region through the refined flow prediction. How to realize the most accurate flow prediction with the lowest cost is a difficult problem faced by each current IP operator, and the current IP operator mainly has two ideas in flow prediction:
firstly, the method comprises the following steps: and (3) performing proportion prediction according to the whole data of the past years, counting the flow values of the past years, and calculating the approximate flow value of the next year, for example, acquiring the flow conditions from 2009 to 2017 of a certain data network outlet through a network management system, and calculating the increase trend of the flow in 2018 according to the change condition of each year and a certain formula. Based on this, the network capacity expansion is carried out. The method has the disadvantages that the data is too general, the influences of user behaviors, regions and the like are ignored, the accuracy of the obtained prediction result cannot be guaranteed, and the waste of capacity expansion is easily caused.
II, secondly: selecting equipment to deploy Deep Packet Inspection equipment (DPI Deep Packet Inspection) on an important equipment level, analyzing user behaviors through Packet capturing analysis of the DPI equipment to perform fine traffic analysis. The predicted result is generally accurate, and the method has the disadvantages of high cost, firstly, DPI equipment needs to be deployed, secondly, the DPI technology is not static and unchangeable, and along with the development of the detection technology, the hiding technology which is abnormally applied is also evolved, for example, the technical cost is correspondingly improved, such as partial encryption of data, characteristic word hiding, detection avoidance through a tunnel technology and the like.
Disclosure of Invention
The embodiment of the invention provides a data network flow prediction method and a data network flow prediction device, which can reflect the whole flow condition of the data network flow and reflect the whole flow growth trend of the data network; the user behavior of each bandwidth under the service type can be embodied, and a basis can be provided for the large speed increase of the broadband.
The embodiment of the invention provides a data network flow prediction method, which comprises the following steps:
acquiring total traffic historical data of a data network, historical online user numbers of different service types and different bandwidths;
based on the established flow prediction model, determining historical average use flow of users with different service types and different bandwidths according to total flow historical data of the data network and historical online user numbers with different service types and different bandwidths;
and predicting the total flow of the data network after capacity expansion based on preset target capacity expansion users of different service types and different bandwidths according to the historical average use flow of the users of different service types and different bandwidths.
The embodiment of the invention also provides a data network flow prediction device, which comprises:
the information acquisition module is used for acquiring total traffic historical data of the data network and historical online user numbers of different service types and different bandwidths;
the flow calculation module is used for determining historical average use flow of users with different service types and different bandwidths according to total flow historical data of the data network and historical online user numbers with different service types and different bandwidths based on a constructed flow prediction model;
and the data network total flow prediction module is used for predicting the data network total flow after capacity expansion based on preset target capacity expansion user numbers of different service types and different bandwidths according to the historical average user flow of the users of different service types and different bandwidths.
The embodiment of the invention also provides computer equipment which comprises a memory, a processor and a computer program which is stored on the memory and can be run on the processor, wherein the processor realizes the data network flow prediction method when executing the computer program.
The embodiment of the invention also provides a computer readable storage medium, and the computer readable storage medium stores a computer program for executing the data network flow prediction method.
In the embodiment of the invention, total flow historical data of a data network, historical online user numbers of different service types and different bandwidths are obtained; based on the established flow prediction model, determining historical average use flow of users with different service types and different bandwidths according to total flow historical data of the data network and historical online user numbers with different service types and different bandwidths; and predicting the total flow of the data network after capacity expansion based on preset target capacity expansion users of different service types and different bandwidths according to the historical average use flow of the users of different service types and different bandwidths. Compared with the prior art, the method can reflect the whole flow condition of the data network flow and reflect the whole flow growth trend of the data network; the user behavior of each bandwidth under the service type can be embodied, and a basis can be provided for the large speed increase of the broadband.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a schematic diagram of a data networking architecture;
fig. 2 is a flow chart (i) of a data network traffic prediction method according to an embodiment of the present invention;
fig. 3 is a flow chart of a data network traffic prediction method according to an embodiment of the present invention (ii);
fig. 4 is a schematic diagram illustrating a comparison result between a bandwidth average traffic value determined by a traffic prediction model according to an embodiment of the present invention and a fluctuation situation of an uplink overall traffic obtained by an actual device;
fig. 5 is a schematic diagram of uplink and downlink flow rates of an IPTV subscriber in the deep east buddha according to an embodiment of the present invention;
fig. 6 is a schematic diagram of a 24-hour change (BRAS sampling rate 5%) of the number of online users of the bead triangle internet and the IPTV provided by the embodiment of the present invention;
FIG. 7 is a schematic diagram of average traffic of bead-triangle Internet users according to an embodiment of the present invention;
fig. 8 is a block diagram (a) of a data network traffic prediction apparatus according to an embodiment of the present invention;
fig. 9 is a block diagram of a data network traffic prediction apparatus according to an embodiment of the present invention (ii).
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The current data networking architecture of each IP operator is roughly shown in fig. 1. In fig. 1, a Broadband Access Server (Broadband Access Server/Broadband Remote Access Server, BAS/BRAS, hereinafter collectively referred to as BRAS) is a new Access gateway for Broadband network applications. It is located at the tandem layer or edge layer of the backbone network, and can complete the data access of the IP/ATM network with user bandwidth (or high speed). The access layer equipment in each data network is connected to the BRAS after being converged, and the data on the BRAS is communicated with the backbone or other data networks through the exit router. The BRAS is connected with the access network downwards, and is connected with the data network outlet, the component flow of each service type in the BRAS is analyzed, the total flow condition of the data network can be analyzed upwards, and the user behavior conditions of various bandwidths of various operators can be analyzed downwards.
Based on the above, the present invention provides a data network traffic prediction method, and the overall idea of the method is to use a BRAS as a unit, and by obtaining the average traffic of users under various bandwidths under various service types of the BRAS, and taking this as basic data, establish a model, such as a region block model, an equipment link model, etc., and analyze data, such as a region block, a traffic condition of an equipment link, and a superimposed user speed-up, etc., according to the model, to form an analysis basis for network capacity expansion and service promotion.
As shown in fig. 2, the data network traffic prediction method includes:
step 201: acquiring total traffic historical data of a data network, historical online user numbers of different service types and different bandwidths;
step 202: based on the established flow prediction model, determining historical average use flow of users with different service types and different bandwidths according to total flow historical data of the data network and historical online user numbers with different service types and different bandwidths;
step 203: and predicting the total flow of the data network after capacity expansion based on preset target capacity expansion users of different service types and different bandwidths according to the historical average use flow of the users of different service types and different bandwidths.
In the embodiment of the invention, the operation and maintenance find that the whole flow of the BRAS has a greater relationship with the public Internet service in the BRAS and IPTV (IPTV, namely interactive network television, is a brand-new technology which utilizes a broadband cable television network, integrates various technologies such as Internet, multimedia, communication and the like, and provides various interactive services including digital television for home users. And for BRAS to be analyzed, the users are classified into bandwidth grades according to the bandwidth, such as: 4M below-4B, 4M-4, 6M-6, 8M-8, 10M-10, 12M-12, 20M above 20B and the like. In this case, the present invention is specifically performed as follows:
step 201 (specific): acquiring total traffic historical data of a data network, historical online user numbers under different bandwidths under public internet services and historical online user numbers under different bandwidths under IPTV according to areas where users are located or different equipment links;
step 202 (specific): based on a constructed regional flow prediction model or a constructed equipment link flow prediction model, determining historical average use flow of users under different bandwidths under the public internet service and historical average use flow of users under different bandwidths under the IPTV according to total flow historical data of the data network, historical online user numbers under different bandwidths under the public internet service and historical online user numbers under different bandwidths under the IPTV;
step 203 (specific): and predicting the total flow of the data network after capacity expansion based on the preset target capacity expansion user number under different bandwidths under the public internet service and the preset target capacity expansion user number under different bandwidths under the IPTV according to the historical average use flow of the users under different bandwidths under the public internet service and the historical average use flow of the users under different bandwidths under the IPTV.
The following describes how the flow prediction model is constructed.
Firstly, a periodic acquisition function of BRAS uplink relay traffic is realized, and the acquisition period is 5 minutes. The acquisition function of the number of the online users of the BRAS branch service type is realized, and the acquisition period is 10 minutes. The acquisition time points of the two samples are overlapped in integral multiples of 10 minutes, and the two samples are corresponding to each other according to the same acquisition time point to be used as an analysis sample.
Before analyzing the actual data, through preliminary analysis, it is considered that:
1) the number of online users of the public internet and the IPTV has a great influence on the BRAS uplink relay traffic (downlink traffic, the traffic mentioned below refers to the downlink traffic of the BRAS uplink relay), so a traffic prediction model is assumed to be:
BARSflow rate＝a×w+b×h+c；
Wherein, BARSFlow rateRepresenting the total flow of the data network; a represents average usage flow of public internet users; w represents the number of online users of the public internet; b represents the average usage flow of IPTV users; h represents the number of online users of the IPTV; and c represents other usage flows.
2) Because the internet access behaviors of users in different areas may have great difference, the correlation between the online user number and the BRAS uplink relay traffic may have great difference in different areas, even different BRASs, and the area model can be established and a plurality of devices can be selected for sampling analysis in the area.
3) The online user number collection that can refine, the bandwidth condition of the active user of data network of gathering in advance (can refer to sampling and gather each bandwidth user flow scheme), and the bandwidth condition is added in each acquisition cycle in the stack, can obtain online user's bandwidth distribution condition, then the model in 1) can become:
BARSflow rate＝a1×w1+a2×w2+a3×w3+b1×h1+b2×h2+b3×h3+c；
Wherein, a1、a2、a3The average usage flow of the public internet users of each bandwidth is represented; w is a1、w2、w3The online user number of each bandwidth of the public Internet is represented; b1、b2、b3The average usage flow of IPTV users of each bandwidth is represented; h is1、h2、h3Representing the number of online users of each bandwidth of the IPTV; and c represents other usage flows.
The model can form a linear equation set through multi-point data, and finally solve a1、a2、a3、b1、b2、b3And average user traffic at various bandwidths.
According to actual data, through preliminary analysis, the following results are found:
1) the flow rate has obvious linear relation with the total online user number, and has similar relation with 163 (referring to public internet) and IPTV user number respectively.
2) By comprehensively analyzing the relationship between the three, the 163 user numbers are mostly in positive linear correlation (a >0), and the IPTV user numbers are in positive correlation (b >0) and in negative correlation (b < 0).
3) The intercept obtained by the linear analysis is a negative number (c <0) regardless of the total number of users or 163 users, which means that the flow rate is negative when the number of users is 0. Perhaps because some users are hanging on the network but do not take up traffic.
The analysis of the superposition correlation can make the predicted flow more accurate.
The following data analysis results for one of the selected devices are shown in table 1:
TABLE 1
In the embodiment of the present invention, the constructed traffic prediction model may include an area traffic prediction model and an equipment link traffic prediction model.
1) Regional flow prediction model
The purpose of the method is to analyze the bandwidth distribution condition of the service types in the region and the average flow condition of each bandwidth. So as to determine the influence of the change of the flow of the film area caused by the acceleration of the broadband in the film area and adjust the capacity expansion index.
The acquisition scheme is based on the original acquisition according to the equipment, the service type and the bandwidth mode, 100 users with various bandwidths of various service types on the independent equipment are randomly selected, and the users who are off-line carry out the user filling with the same service type and the same bandwidth. And the corresponding relationship between the user account information and the film area needs to be acquired.
After data collection is completed, a marketing center and regional branch companies of the user are determined, the marketing center and the regional branch companies are used for data storage, business types, the regional branch companies, the marketing center and bandwidth are collected during collection, and the marketing center mode collection according to the regional branch companies is added when the regional branch companies are used as the upper layers of the marketing center. When selecting the equipment, the equipment is selected according to the distribution of the equipment, and each center approximately selects 2-3 BRASs for collection.
Summarized roughly as shown in table 2:
TABLE 2
| City of land | Regional division | Marketing center | Bandwidth (mbps) | Average rate |
| Buddha mountain | Cis and de | Nanzhuang marketing service center | 4 | 164.19 |
| Buddha mountain | Cis and de | Great marketing service center | 4 | 164.19 |
| Buddha mountain | (South China Sea) | Salt step marketing service center | 4 | 164.19 |
2) Equipment link flow prediction model
The binding relationship between the account and the device port needs to be obtained by the model, some analysis is performed in the early stage, the proportion relationship between the account number bound to the device port in all users is analyzed, and table 3 is one of the analysis:
TABLE 3
| Equipment IP | Total number of users | Bound user populationNumber of | Binding percentage |
| 1xx.xxx.xxx.xxx | 4564 | 3579 | 78.42 |
Sampling analysis is performed on the device with higher binding ratio to obtain ports with more bound users, as shown in table 4:
TABLE 4
| Equipment IP | Port(s) | Port subscriber total number | Port subscriber total ratio |
| 1xx.xxx.xxx.xxx | 1//2 | 479 | 13.38 |
| 1xx.xxx.xxx.xxx | 1//3 | 38 | 1.06 |
| 1xx.xxx.xxx.xxx | 1//4 | 17 | 0.47 |
| 1xx.xxx.xxx.xxx | 1/0/1 | 4 | 0.11 |
| 1xx.xxx.xxx.xxx | 1/0/2 | 1092 | 30.51 |
| 1xx.xxx.xxx.xxx | 1/0/3 | 1065 | 29.76 |
| 1xx.xxx.xxx.xxx | 1/0/4 | 881 | 24.62 |
| 1xx.xxx.xxx.xxx | 14/0/2 | 3 | 0.08 |
The number of the bound users of the device ports is large, the binding percentage reaches 78%, users corresponding to ports 1/0/2, 1/0/3 and 1/0/4 under representative devices can be selected, the users are tracked during collection, the bandwidth distribution condition of the users is analyzed, the average flow of each bandwidth is analyzed according to the bandwidth type, the flow condition of a link corresponding to the output port is counted, and the flow condition of the devices is recurred layer by layer according to the ratio. The method comprises the steps of counting the bandwidth average flow in a link and the bandwidth average flow in equipment, analyzing the influence of bandwidth acceleration on the flow increase of the link, acquiring relevant information of a port link from a network manager, analyzing the current flow utilization rate and the flow utilization rate calculation of a circuit in bandwidth expansion, and then reversing the demand of the expansion on the equipment flow according to the ratio.
In this embodiment of the present invention, as shown in fig. 3, the data network traffic prediction may further include:
step 204: acquiring average user traffic of users of different service types and different bandwidths according to a preset acquisition mode;
step 205: and comparing the acquired user average usage flow of different service types and different bandwidths with the calculated user historical average usage flow of different service types and different bandwidths, and verifying the accuracy of the constructed flow prediction model according to the comparison result.
Specifically, step 104 collects user traffic of each bandwidth according to sampling, specifically collects and acquires user bandwidth of pppoe in the whole province and user city and place, and accesses BRAS information.
The data collected are shown in table 5:
TABLE 5
| Account number | City of land | Bandwidth information (mbps) | Accessing BRAS |
| fsDSL38X8X82@163.gd | Buddha mountain | 1 | XX-XXXX-BRAS-4.MAN.SE800 |
| fsDSL8382X229@163.gd | Buddha mountain | 2 | XX-XXXX-BRAS-4.MAN.SE800 |
| fswiXXnway@163.gd | Buddha mountain | 6 | XX-XXXX-BRAS-4.MAN.SE800 |
| 7X7064266X | Buddha mountain | 6 | XX-XXXX-BRAS-4.MAN.SE800 |
| fsDSLX6X484X7 | Buddha mountain | 6 | XX-XXXX-BRAS-4.MAN.SE800 |
| fsDSL28XX7638 | Buddha mountain | 12 | XX-XXXX-BRAS-4.MAN.SE800 |
| fsDSL13XX690260@163.gd | Buddha mountain | 6 | XX-XXXX-BRAS-4.MAN.SE800 |
| fsDSL1X916X80697 | Buddha mountain | 4 | XX-XXXX-BRAS-4.MAN.SE800 |
| 7X7047XX3381 | Buddha mountain | 4 | XX-XXXX-BRAS-4.MAN.SE800 |
The above data were processed.
Currently, BRAS mainstream equipment is hua shi ME60 and RedBack SE800, and the following collection logic is described by taking the two types of equipment as an example:
ME60：
the right of the login device to execute the relevant command is required.
a. All online users of the corresponding service types are checked by the equipment:
display access-user domain XXX
XXX is the domain corresponding to the service type, and returns:
128fsDSLXX712525@163.gd GE2/0/8.3972 14.157.240.167 f4ec-386f-8461
128 is the user access id.
b. And viewing detailed information of a specific user:
display access-user user-id 125647
the detailed information of the user name, bandwidth, access port and the like can be acquired:
the User name corresponds to User name information, and the Outbound corresponds to bandwidth speed limit information downloaded by the User.
SE800：
The right of the login device to execute the relevant command is required.
a. Find all currently online users that come on through ppoe dial (the user name may not be fully displayed)
sh subscribers all|in pppoe
pppoe 1/1vlan-id 3357:1116pppo fsDXX4362068@163.g 163Mar 3 21:07:41
b. Display find all online users (including non-ppoe users, but with the user name displayed intact)
show sub act all|grep o'-E'^[0-9a-zA-Z]
075785557048
075781826670@163.gd
c. Displaying information of user and vlan interface qos thereof
show sub act username fsDSLXXX7110@163.gd
fsDSL2XX7110@163.gd
Session state Up
qos-metering-policy 12M_MTR(applied)
qos-policing-policy 512K-PLC(applied)
Wherein fsDSL2XX7110@163.gd is account information, and qos-metering-policy 12M _ MTR (applied) is bandwidth information.
For the BRAS to be analyzed, each bandwidth decimation user on the appointed service type of each BRAS performs flow collection, and the users are divided into bandwidth grades according to the bandwidth, such as: the method comprises the following steps of (4M) and (4B), 4M-4, 6M-6, 8M-8, 10M-10, 12M-12, 20M and more than 20B, wherein the granularity is 20B, 10 minutes is one granularity, 100 users are randomly selected in the equipment according to service types (public Internet and IPTV) and the bandwidth of each grade for carrying out traffic collection, users who are off-line in the middle and off-line and on-line again are excluded, when the number of the users is less than 100 during collection, the users are randomly supplemented in the users with the same grade bandwidth, and the user bandwidth data of the original granularity of 10 minutes is collected, and the collection method comprises the following steps:
ME60:
a. view all current online users:
display access-user domain XXX
XXX is the domain corresponding to the service type, and returns:
128fsDSLXX712525@163.gd GE2/0/8.3972 14.157.240.167f4ec-386f-8461
b. view value of user traffic counter:
display access-user user-id 125647
Up packets number(high,low):(0,2961075)
Up bytes number(high,low):(0,957532975)
Down packets number(high,low):(0,3928670)
Down bytes number(high,low):(0,3902569914)
SE800：
a. find all currently online users that come on through ppoe dial (the user name may not be fully displayed)
sh subscribers all|in pppoe
pppoe 1/1vlan-id 3357:1116pppo fsDXX4362068@163.g 163Mar 3 21:07:41
b. Display find all online users (including non-ppoe users, but with the user name displayed intact)
show sub act all|grep o'-E'^[0-9a-zA-Z]
075785557048
075781826670@163.gd
c. Displaying information of user and vlan interface thereof
show sub act username fsDSLXXX7110@163.gd
fsDSL2XX7110@163.gd
Circuit 1/1vlan-id 3357:1103pppoe 6101
d. Searching and displaying rate and other detailed information of single user
how cir count 1/1vlan-id 2407:245pppoe 1307detail
Circuit:1/1vlan-id 2407:245pppoe 1307,Internal id:6/2/14145,Encap:ether-dot1q-tunnel-pppoe-ppp
Packets Bytes
-------------------------------------------------------------------
Receive:693711Receive:142784741
The value of the user flow counter is collected, the difference value between two collection points is used as the total flow between the collection points, the average flow between the collection points is obtained by the total flow/collection point time difference, and the total online number of the current 163 services is collected.
Summarizing data after the original data acquisition is finished:
and summarizing the 10-minute original data according to the time points of the city, the bandwidth type, the service type and the ten-minute granularity, and summarizing the average uplink and downlink flow rate and the maximum uplink and downlink flow rate of the city bandwidth with the ten-minute granularity.
And summarizing the 10-minute summarized data according to the local city, the bandwidth type, the service type and the hour granularity time point, and summarizing the average uplink and downlink flow rate and the maximum uplink and downlink flow rate of the local city bandwidth of the hour granularity.
And summarizing the hour summarized data according to the time points of the city, the bandwidth type, the service type and the daily granularity, and summarizing the average uplink and downlink flow rate and the maximum uplink and downlink flow rate of the city bandwidth of the daily granularity.
For the service type: 163, city, earth: guangzhou, statistical particle size: 10 minutes, bandwidth: the 4M collection is summarized in table 6:
TABLE 6
The following examples illustrate the accuracy of the flow prediction model proposed by the present invention.
1. The flow of each bandwidth user is analyzed through the flow prediction model provided by the invention.
Collecting BARS from 1/6/2013 to 30/6/2013Flow rateThe number of online users of the public internet, the number of online users of the IPTV, and other usage flows are determined by the flow prediction model provided by the present invention, the fluctuation of the uplink overall flow from 2013 in 6 months 1 to 2013 in 6 months 30 is obtained from the actual device, the two are compared, and the comparison result is shown in fig. 4. The fluctuation condition of the bandwidth average flow value obtained by the flow prediction model is consistent with the fluctuation condition of the actual equipment uplink whole flow, and the predicted flow predicted according to historical data is more accurate.
2. And the flow analysis of each bandwidth user is acquired by sampling.
Actual measurement is carried out on the 24-hour traffic of public Internet users and IPTV users in Guangdong Dongfu 4 in 12 months in xx years, 30 BRAS are randomly selected to measure the uplink/downlink average flow rate of 2 types of users, 1000 IPTV users exist, and the total number of the public users is nearly 30 ten thousand.
Measuring the flow rate of the public internet users: the traffic of 30 BRAS of 4 BRAS is actually measured in 24 hours, the sampling interval is 30 minutes (the sampling rate is 5 percent, the coverage is 19 ten thousand users), the average downstream rate of the peak value of the internet user in the evening is about 508kbps, and the average upstream rate of the peak value is about 182kbps (the IPTV traffic is deducted);
measuring the flow rate of the IPTV user: monitoring accounts of 1000 IPTV users in 4 BRAS for 12 hours, wherein the sampling interval is 15 minutes, the downlink rate of the IPTV users at night is basically stabilized at 1.6Mbps (3 times of that of Internet users), and the average uplink rate is 70 kbps;
the number of IPTV users vs the number of Internet online users is 1:10, and the number of high-definition users vs is 1:10 (far lower than that of common-definition users).
Fig. 5 is a schematic diagram of uplink and downlink flow rates of a wide and deep east Buddha IPTV user (where the abscissa represents time and the ordinate represents flow rate), fig. 6 is a schematic diagram of a 24-hour change (BRAS sampling rate 5%) between a bead-triangle internet and an IPTV online user number (where the internet online user number is represented by a left ordinate value and the IPTV online user number is represented by a right ordinate value and the abscissa represents time), and fig. 7 is a schematic diagram of average flow rates of the bead-triangle internet users (where the BARS downlink flow rate is represented by a left ordinate value, the downlink average flow rate and the uplink average flow rate are represented by a right ordinate value and the abscissa represents time). As can be seen from fig. 5 to fig. 7, in the uplink traffic analysis of the BRAS device, the fluctuation situation of the overall traffic of the BRAS device substantially matches the sampling analysis result.
The results analyzed by the two methods can basically reflect the actual flow conditions of the existing network, and the data have reference value.
Based on the same inventive concept, the embodiment of the present invention further provides a data network traffic prediction apparatus, as described in the following embodiments. Because the principle of solving the problems of the data network flow prediction device is similar to that of the data network flow prediction method, the implementation of the data network flow prediction device can refer to the implementation of the data network flow prediction method, and repeated parts are not described again. As used hereinafter, the term "unit" or "module" may be a combination of software and/or hardware that implements a predetermined function. Although the means described in the embodiments below are preferably implemented in software, an implementation in hardware, or a combination of software and hardware is also possible and contemplated.
Fig. 8 is a block diagram of a data network traffic prediction apparatus according to an embodiment of the present invention, and as shown in fig. 8, the apparatus includes:
an information obtaining module 801, configured to obtain total traffic historical data of a data network, historical online user numbers of different service types and different bandwidths;
a traffic calculation module 802, configured to determine historical average usage traffic of users of different service types and under different bandwidths, according to total traffic historical data of the data network, historical online users of different service types and under different bandwidths, based on a constructed traffic prediction model;
and the total data network traffic prediction module 803 is configured to predict total data network traffic after capacity expansion based on preset target capacity expansion user numbers of different service types and different bandwidths according to historical average user traffic of users of different service types and different bandwidths.
This structure will be explained below.
In the embodiment of the invention, the service types comprise public internet service and interactive network television IPTV;
the information obtaining module 801 is specifically configured to:
acquiring total traffic historical data of a data network, historical online user numbers under different bandwidths under public internet services and historical online user numbers under different bandwidths under IPTV according to areas where users are located or different equipment links;
the flow calculation module 802 is specifically configured to:
based on a constructed regional flow prediction model or a constructed equipment link measurement model, determining historical average use flow of users under different bandwidths under the public internet service and historical average use flow of users under different bandwidths under the IPTV according to total flow historical data of the data network, historical online user numbers under different bandwidths under the public internet service and historical online user numbers under different bandwidths under the IPTV;
the total data network traffic prediction module 803 is specifically configured to:
and predicting the total flow of the data network after capacity expansion based on the preset target capacity expansion user number under different bandwidths under the public internet service and the preset target capacity expansion user number under different bandwidths under the IPTV according to the historical average use flow of the users under different bandwidths under the public internet service and the historical average use flow of the users under different bandwidths under the IPTV.
In the embodiment of the present invention, the constructed flow prediction model has the following specific form:
BARSflow rate＝a1×w1+a2×w2+a3×w3+b1×h1+b2×h2+b3×h3+c；
Wherein, a1、a2、a3The average usage flow of the public internet users of each bandwidth is represented; w is a1、w2、w3The online user number of each bandwidth of the public Internet is represented; b1、b2、b3The average usage flow of IPTV users of each bandwidth is represented; h is1、h2、h3Representing the number of online users of each bandwidth of the IPTV; and c represents other usage flows.
In this embodiment of the present invention, the information obtaining module 801 is further configured to:
acquiring average user traffic of users of different service types and different bandwidths according to a preset acquisition mode;
as shown in fig. 9, the data network traffic prediction apparatus further includes:
the verification module 804 is configured to compare the acquired average user traffic of different service types and different bandwidths with the calculated historical average user traffic of users of different service types and different bandwidths, and verify the accuracy of the constructed traffic prediction model according to the comparison result.
The embodiment of the invention also provides computer equipment which comprises a memory, a processor and a computer program which is stored on the memory and can be run on the processor, wherein the processor realizes the data network flow prediction method when executing the computer program.
The embodiment of the invention also provides a computer readable storage medium, and the computer readable storage medium stores a computer program for executing the data network flow prediction method.
In summary, the data network traffic prediction method and apparatus provided by the present invention have the following beneficial effects;
historical data of total traffic of a data network, historical online user numbers of different service types and different bandwidths are obtained; based on the established flow prediction model, determining historical average use flow of users with different service types and different bandwidths according to total flow historical data of the data network and historical online user numbers with different service types and different bandwidths; and predicting the total flow of the data network after capacity expansion based on preset target capacity expansion users of different service types and different bandwidths according to the historical average use flow of the users of different service types and different bandwidths. Compared with the prior art, the method can reflect the whole flow condition of the data network flow and reflect the whole flow growth trend of the data network; the user behavior of each bandwidth under the service type can be embodied, and a basis can be provided for the large speed increase of the broadband.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The above description is only a preferred embodiment of the present invention, and is not intended to limit the present invention, and various modifications and changes may be made to the embodiment of the present invention by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. A data network flow prediction method is characterized by comprising the following steps:
acquiring total traffic historical data of a data network, historical online user numbers of different service types and different bandwidths;
based on the established flow prediction model, determining historical average use flow of users with different service types and different bandwidths according to total flow historical data of the data network and historical online user numbers with different service types and different bandwidths;
predicting the total flow of the data network after capacity expansion based on preset target capacity expansion users of different service types and different bandwidths according to the historical average use flow of the users of different service types and different bandwidths;
the constructed traffic prediction model comprises a constructed regional traffic prediction model or a constructed equipment link traffic prediction model;
the constructed equipment link flow prediction model is constructed as follows:
acquiring the binding relationship between the account and the equipment port, and analyzing the binding proportion of the account bound with the equipment port in all users;
sampling analysis is carried out on equipment with a binding ratio exceeding a preset binding ratio threshold, users corresponding to ports with the binding user number exceeding the preset port user total number of equipment ports are selected, the bandwidth distribution condition of the users is analyzed, average flow of different bandwidths is analyzed according to bandwidth types, the flow condition of a link corresponding to an output port is counted, the flow condition of the equipment is recurred layer by layer according to the binding ratio, the bandwidth average flow in the link and the bandwidth average flow in the equipment are counted, the influence of bandwidth acceleration on the flow increase of the link is analyzed, relevant information of the port link is obtained from a network manager, the current flow utilization ratio is analyzed, bandwidth expansion is calculated according to the flow utilization ratio, and the requirement of expansion on the equipment flow is reversed according to the binding ratio.
2. The data network traffic prediction method of claim 1 wherein the traffic types include public internet traffic and interactive network television, IPTV;
acquiring total traffic historical data of a data network, historical online user numbers of different service types and different bandwidths, wherein the historical online user numbers comprise:
acquiring total traffic historical data of a data network, historical online user numbers under different bandwidths under public internet services and historical online user numbers under different bandwidths under IPTV according to areas where users are located or different equipment links;
based on the established flow prediction model, determining historical average user flow of users with different service types and different bandwidths according to total flow historical data of the data network and historical online user numbers with different service types and different bandwidths, wherein the historical average user flow of the users with different service types and different bandwidths comprises the following steps:
based on a constructed regional flow prediction model or a constructed equipment link flow prediction model, determining historical average use flow of users under different bandwidths under the public internet service and historical average use flow of users under different bandwidths under the IPTV according to total flow historical data of the data network, historical online user numbers under different bandwidths under the public internet service and historical online user numbers under different bandwidths under the IPTV;
predicting the total flow of the data network after capacity expansion based on the preset target capacity expansion user number under different service types and different bandwidths according to the historical average use flow of the users under different service types and different bandwidths, wherein the method comprises the following steps:
and predicting the total flow of the data network after capacity expansion based on the preset target capacity expansion user number under different bandwidths under the public internet service and the preset target capacity expansion user number under different bandwidths under the IPTV according to the historical average use flow of the users under different bandwidths under the public internet service and the historical average use flow of the users under different bandwidths under the IPTV.
3. The data network traffic prediction method of claim 2, characterized in that the constructed traffic prediction model is in the following specific form:
BARSflow rate＝a1×w1+a2×w2+a3×w3+b1×h1+b2×h2+b3×h3+c；
Wherein, BARSFlow rateRepresenting the total flow of the data network; a is1、a2、a3The average usage flow of the public internet users of each bandwidth is represented; w is a1、w2、w3The online user number of each bandwidth of the public Internet is represented; b1、b2、b3The average usage flow of IPTV users of each bandwidth is represented; h is1、h2、h3Representing the number of online users of each bandwidth of the IPTV; and c represents other usage flows.
4. The data network traffic prediction method of claim 1, further comprising:
acquiring average user traffic of users of different service types and different bandwidths according to a preset acquisition mode;
and comparing the acquired user average usage flow of different service types and different bandwidths with the calculated user historical average usage flow of different service types and different bandwidths, and verifying the accuracy of the constructed flow prediction model according to the comparison result.
5. A data network traffic prediction apparatus, comprising:
the information acquisition module is used for acquiring total traffic historical data of the data network and historical online user numbers of different service types and different bandwidths;
the flow calculation module is used for determining historical average use flow of users with different service types and different bandwidths according to total flow historical data of the data network and historical online user numbers with different service types and different bandwidths based on a constructed flow prediction model;
the data network total flow prediction module is used for predicting the total flow of the data network after capacity expansion based on preset target capacity expansion user numbers of different service types and different bandwidths according to historical average use flows of users of different service types and different bandwidths;
the constructed traffic prediction model comprises a constructed regional traffic prediction model or a constructed equipment link traffic prediction model;
the constructed equipment link flow prediction model is constructed as follows:
acquiring the binding relationship between the account and the equipment port, and analyzing the binding proportion of the account bound with the equipment port in all users;
sampling analysis is carried out on equipment with a binding ratio exceeding a preset binding ratio threshold, users corresponding to ports with the binding user number exceeding the preset port user total number of equipment ports are selected, the bandwidth distribution condition of the users is analyzed, average flow of different bandwidths is analyzed according to bandwidth types, the flow condition of a link corresponding to an output port is counted, the flow condition of the equipment is recurred layer by layer according to the binding ratio, the bandwidth average flow in the link and the bandwidth average flow in the equipment are counted, the influence of bandwidth acceleration on the flow increase of the link is analyzed, relevant information of the port link is obtained from a network manager, the current flow utilization ratio is analyzed, bandwidth expansion is calculated according to the flow utilization ratio, and the requirement of expansion on the equipment flow is reversed according to the binding ratio.
6. The data network traffic prediction device of claim 5, wherein the service types include public internet service and interactive network television, IPTV;
the information acquisition module is specifically configured to:
acquiring total traffic historical data of a data network, historical online user numbers under different bandwidths under public internet services and historical online user numbers under different bandwidths under IPTV according to areas where users are located or different equipment links;
the flow calculation module is specifically configured to:
based on a constructed regional flow prediction model or a constructed equipment link flow prediction model, determining historical average use flow of users under different bandwidths under the public internet service and historical average use flow of users under different bandwidths under the IPTV according to total flow historical data of the data network, historical online user numbers under different bandwidths under the public internet service and historical online user numbers under different bandwidths under the IPTV;
the data network total flow prediction module is specifically configured to:
and predicting the total flow of the data network after capacity expansion based on the preset target capacity expansion user number under different bandwidths under the public internet service and the preset target capacity expansion user number under different bandwidths under the IPTV according to the historical average use flow of the users under different bandwidths under the public internet service and the historical average use flow of the users under different bandwidths under the IPTV.
7. The data network traffic prediction device of claim 6, wherein the constructed traffic prediction model is in the following specific form:
BARSflow rate＝a1×w1+a2×w2+a3×w3+b1×h1+b2×h2+b3×h3+c；
Wherein, BARSFlow rateRepresenting the total flow of the data network; a is1、a2、a3The average usage flow of the public internet users of each bandwidth is represented; w is a1、w2、w3The online user number of each bandwidth of the public Internet is represented; b1、b2、b3The average usage flow of IPTV users of each bandwidth is represented; h is1、h2、h3Representing the number of online users of each bandwidth of the IPTV; and c represents other usage flows.
8. The data network traffic prediction device of claim 5, wherein the information acquisition module is further configured to:
acquiring average user traffic of users of different service types and different bandwidths according to a preset acquisition mode;
further comprising:
and the verification module is used for comparing the acquired average user traffic of different service types and different bandwidths with the calculated historical average user traffic of the users of different service types and different bandwidths, and verifying the accuracy of the constructed traffic prediction model according to the comparison result.
9. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the data network traffic prediction method of any one of claims 1 to 4 when executing the computer program.
10. A computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program for executing the data network traffic prediction method according to any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811526034.9A CN109495317B (en) | 2018-12-13 | 2018-12-13 | Data network flow prediction method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811526034.9A CN109495317B (en) | 2018-12-13 | 2018-12-13 | Data network flow prediction method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109495317A CN109495317A (en) | 2019-03-19 |
| CN109495317B true CN109495317B (en) | 2022-01-18 |
Family
ID=65710083
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811526034.9A Active CN109495317B (en) | 2018-12-13 | 2018-12-13 | Data network flow prediction method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109495317B (en) |
Families Citing this family (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109978627B (en) * | 2019-03-29 | 2023-08-08 | 电子科技大学中山学院 | Modeling method for big data of broadband access network user surfing behavior |
| CN110300014A (en) * | 2019-04-17 | 2019-10-01 | 中国联合网络通信集团有限公司 | A kind of flow analysis method and device |
| CN111865628B (en) * | 2019-04-25 | 2023-04-07 | 中国移动通信集团河北有限公司 | Statistical system, method, server and storage medium for influencing user by home wide fault |
| CN112187498B (en) * | 2019-07-03 | 2022-09-06 | 中国电信股份有限公司 | Bypass protection method, device and system thereof and Deep Packet Inspection (DPI) system |
| CN110380904B (en) * | 2019-07-23 | 2021-09-21 | 腾讯科技（深圳）有限公司 | Bandwidth allocation method and device, electronic equipment and storage medium |
| CN110932993A (en) * | 2019-11-21 | 2020-03-27 | 中盈优创资讯科技有限公司 | Bandwidth speed regulation management method and device |
| CN111212337B (en) * | 2019-12-27 | 2022-02-25 | 中国联合网络通信集团有限公司 | Port classification method and device |
| CN114531706B (en) * | 2020-10-30 | 2023-11-10 | 中国移动通信集团江西有限公司 | Regional user traffic monitoring method and device, storage medium and computer equipment |
| CN114449569B (en) * | 2020-11-02 | 2024-01-16 | 中国移动通信集团广东有限公司 | User traffic usage processing method, network equipment and service processing system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103200114A (en) * | 2013-04-18 | 2013-07-10 | 湖北邮电规划设计有限公司 | Metropolitan area network planning method |
| CN105024875A (en) * | 2014-04-30 | 2015-11-04 | 中国电信股份有限公司 | Method and system for measuring network flow rate of broadband classified users |
| CN105407494A (en) * | 2015-10-23 | 2016-03-16 | 中国联合网络通信集团有限公司 | Network capacity expansion method and apparatus |
| CN108259263A (en) * | 2017-12-01 | 2018-07-06 | 国家电网公司 | Data analysing method, apparatus and system |
-
2018
- 2018-12-13 CN CN201811526034.9A patent/CN109495317B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103200114A (en) * | 2013-04-18 | 2013-07-10 | 湖北邮电规划设计有限公司 | Metropolitan area network planning method |
| CN105024875A (en) * | 2014-04-30 | 2015-11-04 | 中国电信股份有限公司 | Method and system for measuring network flow rate of broadband classified users |
| CN105407494A (en) * | 2015-10-23 | 2016-03-16 | 中国联合网络通信集团有限公司 | Network capacity expansion method and apparatus |
| CN108259263A (en) * | 2017-12-01 | 2018-07-06 | 国家电网公司 | Data analysing method, apparatus and system |
Non-Patent Citations (1)
| Title |
|---|
| 城域数据网方案设计及优化;谢美德;《中国优秀硕士学位论文全文数据库信息科技辑》;20150515;第29-33页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109495317A (en) | 2019-03-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109495317B (en) | Data network flow prediction method and device | |
| CN106790050B (en) | A kind of anomalous traffic detection method and detection system | |
| Shafiq et al. | A first look at cellular machine-to-machine traffic: large scale measurement and characterization | |
| CN106126391A (en) | System monitoring method and apparatus | |
| CN106203679A (en) | A kind of customer loss Forecasting Methodology and system | |
| CN106651458A (en) | Advertisement anti-cheating method and device | |
| CN101741608B (en) | Traffic characteristic-based P2P application identification system and method | |
| CN112217725B (en) | Delay optimization method based on edge calculation | |
| CN110147803A (en) | Customer churn early-warning processing method and device | |
| CN105429792B (en) | User behavior flow acquisition methods and device, user behavior analysis method and system | |
| US20200213900A1 (en) | Data congestion management system and method | |
| CN107566236A (en) | For obtaining the methods, devices and systems in client's special line teleservice path | |
| CN106780125A (en) | A kind of acquisition abnormity urgency level computational methods based on monthly power consumption | |
| CN105490865A (en) | Method and device for implementing flow analysis | |
| CN103716211B (en) | The data traffic measuring method of the network terminal | |
| Ridoux et al. | Seeing the difference in IP traffic: Wireless versus wireline | |
| CN109144837B (en) | User behavior pattern recognition method supporting accurate service push | |
| CN107846614B (en) | Video traffic scheduling method and device and electronic equipment | |
| CN105376105A (en) | Internet traffic modeling method based on time-sliding window | |
| CN108322354A (en) | One kind is escaped the recognition methods of flow account and device | |
| CN106257867A (en) | A kind of business recognition method encrypting flow and device | |
| Yeom et al. | LSTM-based collaborative source-side DDoS attack detection | |
| CN101986608A (en) | Method for evaluating heterogeneous overlay network load balance degree | |
| Balafif et al. | Analysis of Computer Network Performance on Communication and Informatics Office of West Sumbawa Regency Using Quality of Service Method | |
| CN103279816A (en) | Active window-based terminal work efficiency statistical method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |